EE: Hello, welcome to this ecancer virtual discussion; we’re hoping soon to come back to the round table discussions as well. Let’s get to it. I think Twitter went crazy the moment we heard back-to-back the data from phase III MAGNITUDE and PROpel trials, so similar and so different in the same domain of metastatic castration resistant prostate cancer. It really traumatised some of us and you will, of course, understand why, knowing that both Gerhardt Attard and myself are so dedicated to biomarker development that we feel we may be regressing. But I will let Axel, as the unbiased independent reviewer, to give us his thoughts on the two trials.

AM: Eleni, thanks a lot and, as you said, it was a very exciting prostate oral session with those both trials which caused not only on social media but also within the personal communication we had afterwards some discussion. Since they were not that easy to understand I will try to do my best to summarise them now. The first one was the MAGNITUDE, it’s a phase III trial, the first results of niraparib, a PARP inhibitor, with abiraterone acetate and prednisone as first-line therapy for patients with metastatic castration resistant prostate cancer with and without HRR gene alterations. So we are really having precision medicine here included in uro-oncology which I think is great, as you just said.

The trial design was pretty smart, randomising one-to-one to niraparib plus abiraterone and placebo plus abiraterone, as said in the mCRPC setting. Of note, it is important, four months of abiraterone previous therapy were allowed. Then 400 men were in the HRR biomarker positive group and 600 men in the HRR biomarker negative group and the primary endpoint was radiographic progression free survival and the biomarker status.

As one would probably expect, or have thought of, the better group was the biomarker positive group. So the HRR negative the composite progression endpoint, radiographic or PSA progression, there was no difference in the HRR negative patients. When looking at the positive group, the rPFS was significantly better – 16.6 months niraparib plus abiraterone compared to placebo plus abiraterone 10.9 months for the BRCA1 or 2 altered patients.

There was still a difference in all HRR positive patients – 16.5 months for the niraparib and abiraterone group
and placebo/abiraterone group 13.7 months.

So this, just, in a nutshell the first MAGNITUDE trial, showing that a combination of niraparib, a PARP inhibitor, with abiraterone in the altered patients, HRR positive patients, makes sense and prolongs progression free survival. Overall survival there was no statistically significant signal as of yet. The second trial, the PROpel trial, also a large phase III trial of olaparib, also a PARP inhibitor, and abiraterone – so the same co-drug – versus placebo and abiraterone as first-line therapy for patients with metastatic castration resistant prostate cancer. So also in the mCRPC setting.

This was an all comer trial so regardless of the mutation, which was brave initially but retrospectively the mutation was investigated. So we know the data, not only from liquid biopsy but also from a great portion of tissue which wasn’t mentioned initially in the abstract. Here in a one-to-one patient those mCRPC patients were randomised to olaparib and abiraterone plus prednisone or placebo plus abiraterone and prednisone. The primary endpoint was
radiographic progression free survival or progression or death. So, as we can expect in a large phase III trial, patient characteristics were well balanced among the groups and the primary endpoint, according to investigator judgement, was met, positive with a hazard ratio of 0.66 for the combination of olaparib and abiraterone compared to placebo and abiraterone. These findings were also positive when looking at the subgroups – this olaparib/abiraterone PFS was better when looking at subgroups which was also interesting in the non-HRR mutated with a hazard ratio of still 0.76. So a lot of discussion whether or not we need the testing and this is probably something we will discuss.

So trials which seem very similar but are not from the trial design. Different mode of action in the drugs used and different comparator or additional drug with abiraterone. So exciting times to discuss. Thanks, and I hope you could have followed me on those two not so easy to understand trials.

EE: Thank you Axel for this great summary. I have to fully disclose, I am an investigator of the MAGNITUDE trial. I want to comment briefly that the two trials were even culturally different because in the HRR negative arm of the PROpel we essentially did not fully accrue, we stopped short at that 200 mark out of 600 because we did that futility analysis that was pre-planned that showed no difference in benefit at that time point. We can go back to the
discussion, I’ll get your thoughts on whether you think this was too soon. Noel Clarke, who did lead the trial, the phase II trial, that unveiled a signal to suggest that there may be a combination benefit with olaparib plus abiraterone in mCRPC, not just for DDR positivity but potentially for DDR negativity. Your work, and I want you to talk a little bit about that and give us your thoughts on the two trials, changed the design of a lot of ongoing trials, by the way. You know that, right?

NC: Yes, thanks Eleni. It’s an interesting story, as you’ve already said, and it started with the basic translational research work that showed that it was possible to get a synergistic effect by adding abiraterone to a PARP inhibitor like olaparib. That was the basis for what was known as study 8 which was a phase II study of 142 patients randomised. But the important thing that we wanted to show in that trial was that there was effect across a section of presenting prostate cancer patients and these were all pretty heavily pre-treated, they virtually all had docetaxel
and had failed first line therapy.

What we found and published was that there was an effect, that it was beneficial and that it was independent of HRR mutational status. So we then sat down and planned the PROpel trial and the philosophy there was to bring the drug combination a little further forward in the disease pathway. So we were aiming to give the drug as first-line therapy, before docetaxel, in patients who had not previously received a novel hormone agent such as abiraterone or enzalutamide.

We planned that at scale in an international setting; the original target was about 750 and we weighed in at just short of 800 patients randomised either way. The essence was that patients had to have no prior DDR test at all. They went into the trial, randomised straight away, but they did have their HRR mutational status measured for retrospective analysis.

The primary endpoint we aimed at was radiological progression free survival with standard imaging. There were two components to that, one was the individual clinicians’ which assessed the radiology but then there was a separate assessment to quality check that independently, which was done, essentially, by experts. At the end of the study we found that actually the experts were slightly aggressive in determining radiological progression free survival than the clinicians in the individual departments.

So, to our pleasant surprise, the findings of study 8 were borne out and that was that there was no difference in the different DDR types; that patients who were HRR mutated got an effect but, equally, those patients who had no HRR mutation also got an affect and it was right across the piece for radiological progression free survival. We had other endpoints, the sort of conventional things like time to second treatment and treatment failure, with the arrows all going in the same direction, in other words, benefit from the combination. The overall survival is immature. We had just under a third of the patients have had an event and that was at 24 months but the survival curves were starting to separate quite significantly, not statistically significantly, but you could see the division in the actuarial curves at the 24 month point. We will follow up that. I expect probably third or fourth quarter of 2022 we’ll have the overall survival which we will present again and that will be more difficulty.

EE: Looking forward to changing my mind thanks to your work. But I will turn to Gerhardt for that. Beyond everything, we are scientists, we are clinicians. Gerhardt, you’ve done tremendous translational research and you are a basic scientist. At the end of the day, we’ll need to be convinced by the data but I wanted to get your thoughts. We all knew that phase II trial, it was very impressive, what do you think about the mechanistic aspect of the design and where do you think we are going? We got the first positive trial for prostate with niraparib but niraparib is approved in ovarian cancer; olaparib, we all know, started its course in 2007 and has done wonderfully also in other domains. What are your thoughts with regard to the role?

GA: I’m going to give a view through my biomarker tinted glasses and then maybe we can handover back to Noel to say what he thinks about this interpretation. Because there has been a lot of talk about how different they are, both the trials, the results, the population, but they don’t look too different to me. What is clear to me, or what is suggested to me by the two presentations, is that PARP inhibition works for the HRR mutated. In magnitude there’s evidence that it works better in BRCA; that’s probably not as obvious in PROpel, the numbers may not be there. One can argue about over-interpreting forest plots in PROpel but my read of the forest plot in PROpel and looking at the magnitude data would suggest that the efficacy of PARP inhibition is greater in HRR mutated and probably in BRCA mutated. That’s in terms of rPFS. The big question now is will this translate into an improvement in overall survival? So this really is an early versus late question for the HRR mutant and specifically the BRCA mutant.

The second question is whether there is efficacy in the HRR wildtype. There are a couple of uncertainties in the data. The first you alluded to, Eleni, and [?? 13:42] did as well – that the futility was in magnitude, so the futility was met in the biomarker negative patients, primarily based on PSA endpoints and in a relatively smaller group of patients. So maybe the jury is out there on whether there is some activity or there is some efficacy. Similarly in PROpel there has been a lot of discussion about is the effect in the non-HRR mutant caused by hidden HRR
mutants that were missed by sequencing but I don’t think that could explain that effect. We know what the prevalence is and doing the numbers doesn’t explain that.

So, to summarise, I interpret these as I think the effect is greater in the HRR mutant and I’d be really keen to see more clearly what the benefit is in the BRCA mutant patients that we’ve seen repeatedly previously are the ones that derive greatest benefit. Then the second question, and I think this is really exciting from PROpel, is that there is also efficacy in the non-HRR mutant.

So the one difference of opinion that I have is that I don’t think it’s a case of regardless. I think HRR mutation status does make a difference to efficacy but there is really exciting evidence here that there could also be impact on rPFS, or there is impact on rPFS, and hopefully on OS as well in the HRR non-mutant.

EE: Thank you for that. We don’t have time to go in a lot of detail but I want to hear from Noel because I cut you off. So are you a believer of this agnostic approach or do you… I’m going to correct that and say believe that there is a chance that within the wildtype by HRR there is a potentially BRCA2 behaving subset and we just need to unveil it, identify a signature, identify those BRCA behaving tumours? It’s a very important question because it has come up in a lot of the discussions that the BRCAness is not just defined by what we are looking at right now.

NC: Well, I do believe, and the evidence supports for PROpel, that the HRR wildtype do have an effect. It’s also true to say that in the two trials, PROpel and MAGNITUDE, there were more BRCA patients in MAGNITUDE than PROpel. The other thing to say is that it’s probably more effective in BRCA2 but there were some other key things.

So clearly there’s a lot of headscratching going on as to why MAGNITUDE was different to PROpel and there are a couple of things which need, perhaps, highlighting. One, in the study design for MAGNITUDE the use of abiraterone was allowed early on in a proportion of patients and that has a programming effect on the androgen receptor and how it behaves and that might have been critical in how things panned out in the DDR mutated group and in the non-mutated group. The second thing is that the toxicity of niraparib is higher than olaparib and it’s the haematological toxicity. So the dose reduction and the dose at a therapeutic level at the cellular level is probably a little bit lower than it might be because you have to scale it down to stop the haematological toxicity. Now, there are other factors which we probably don’t have time to go into today, but…

EE: No, absolutely. And thank you for bringing up the abiraterone point. I was hoping that the discussant would have picked it up. I want to go back to you, Axel, because I want to give you the stage now to present a trial that you led, the PRESIDE trial, and I know that Gerhardt is a big co-lead investigator on the trial. I want to hear from both of you the great data that suggests that androgen signalling addiction is there in the tumours, some of the tumours, and how you will proceed in unveiling those subsets. So back to you.

AM: Eleni, thanks. I think the PRESIDE trial, which was a phase III trial and I was very honoured to present the data on behalf of the co-investigators, showed a significant progression free survival improvement when men with chemotherapy naïve metastatic castration resistant prostate cancer were treated with a regimen of continuous enzalutamide versus placebo in combination with chemotherapy with docetaxel. The progression free survival improvement was observed and the median PFS was 9.5 months for those treated with continuous enzalutamide
upon progression and then combined with docetaxel, versus 8.2 months in the placebo arm.

This PFS also translated, there was also a delayed time to prostate specific antigen, PSA, progression at 8.4 months in the enzalutamide group compared to 6.2 months in the placebo group. Of note, I think it is important that we are speaking on the mCRPC situation here. We cannot expect PEACE-1 results or ARASENS results in the non-mHSPC situation.

So the trial was designed, and Gerhardt can correct me there, almost 12-15 years ago when we weren’t talking about therapy intensification, neither in standard treatment was in mHSPC ADT and in mCRPC we had the PREVAIL trial, the COUGAR trial. So there was a very interesting approach at that time and now we have the results at this point. Looking at the safety results from the PRESIDE trial it doesn’t add a lot of toxicity to have a dual hormone blockade upon progression.

Surely we need to discuss the clinical value of this [?? 20:08] a couple of PFS months’ benefit and, as of yet, no overall survival benefit. But this is something hypothesis generating and we are in a new era now of prostate cancer. So what do we do after ARASENS progression? What do we do after PEACE-1 progression or after TITAN progression when we combine ADT with apalutamide? Should we continue apalutamide and add on to that what we just discussed – olaparib in BRCA positive or even BRCA negative patients or looking at the situation, then
mode of action changed to cabazitaxel and to docetaxel and those other options? We have lutetium therapy where we saw a lot of abstracts at ASCO GU. So some interesting pieces of puzzle for the whole picture of advanced prostate cancer. Thank you.

EE: I was sure that the German would bring in lutetium to the discussion. Joking aside, great work.
Gerhardt, are you hoping to unveil that subset that may merit or do you think that this is
something that may help more patients? Because we did not see a huge clinical benefit
arguably.
GA: Yes, it’s the perfect biomarker experiment because we can see evidence of efficacy but
probably not great enough to use enzalutamide for everyone. So unlike some of the work we’re
doing with much more effective treatments, here the difference in the subgroups that we hope
we will see is… let’s say a subgroup with a hazard ratio of 0.55 versus another subgroup with
0.9, for example. That’s the sort of hypothesis generating data we can generate.
I guess to clarify, and kudos to Astellas for collecting plasma and collecting ARV7 or CTCs for
ARV7 analysis from their study, and they’ve partnered with my group and with others to identify
those two subgroups. As Axel said, this is hypothesis generating but if we identify half of
patients who have a hazard ratio of 0.5 when you continue enzalutamide on docetaxel with
really minimal toxicity, there will be some financial toxicity, but I think that will become
important.
The last thing to say is since we designed the original abiraterone and enzalutamide trials
we’ve always said continue beyond PSA progression but we’ve never proven that makes a
difference. This is probably one of the first studies where we’ve continued beyond progression
and shown a better outcome, of course in combination with docetaxel. But I think it’s the first
one to prove that continuing longer might help for some patients.
EE: That’s a great point, again not picked up by the discussant. Thank you for that, I didn’t hear it, even though she did a great job, Elizabeth, in discussing both trials. I wanted, in the queue of that, we presented with Gerhardt and the other co-authors the ctDNA analysis of the ACIS trial.

Again, this was a trial that had met the primary endpoint of rPFS by a margin of 6 months, just to remind you. Published in Lancet Oncology comparing abiraterone plus apalutamide versus apalutamide in mCRPC L1. We know it has benefit in the long term, again, making it a good biomarker study. We’re not fully there yet but we showed some data that actually could be linked to what you’re looking at, suggesting that there are genomic events that we can pick up, either at baseline or upon progression, maybe incurred by the treatment and may predict lack of response to subsequent androgen signalling inhibition.

I’m sure you’re going to look at that, are you going beyond that? Is there an opportunity to combine datsets? What do you think, Gerhardt?

GA: That would be cool. We’ve succeeded previously in working with both Janssen and Astellas together so, yes, I think that’s a good idea, Eleni.

EE: So now it’s going to be about bringing Janssen and Pfizer together, that’s going to be another thing, and then AstraZeneca potentially, let’s see. So, Noel, we want to get your thoughts in that route of developing biomarker strategies. Do you think it’s actionable potentially in the future for urologists? Because you are coming into the forefront of research as you deserve, all of you. We saw you, Axel, Fred [?]. But do you think it’s going to be actionable or will urologists say, ‘Okay, I’m going to go for the easy way out. Just continue my enzalutamide, my
apalutamide and we’ll decide if I transition to an abiraterone plus olaparib.’ Just a thought, and we don’t have data for that transition.

NC: No, we don’t but the use of these agents internationally varies quite considerably country to country. So in some countries, and where Gerhardt and I work, it’s often quite difficult for urologists to prescribe drugs which are coming through in large-scale testing. Whereas in countries where Axel works, for example, urologists can prescribe even cytotoxic drugs.

But getting to the nub of the question, I think that urologists have to be given access to this kind of data. They have to understand what’s going on with their patients because they are actually the gatekeepers in most countries. I like to think that although urologists are fairly simple people, generally speaking, this is actually quite a simple problem when you break it down into its component parts. So I think we could sub-stratify patients in a perfectly reasonable way which would enable, let me say clinicians rather than surgeons and non-surgeons, clinicians
dealing with prostate cancer patients to give the patients the best opportunity that they should have and it should be their right to. Funding it is another issue.

EE: Well, definitely not simple and definitely very efficient. I want Axel to just close it all in with referring to quality of life patient reported outcomes. We had a nice analysis bringing together SPARTAN and TITAN – what are your thoughts on the association with PSA outcomes for these men? Do you think there’s value there? How would you consider it in the clinic?

AM: Eleni, thanks for switching over to the urologist again when it comes to quality of life. Sure, we like hormone-hormone combinations and that’s what was looked at by Eric Small and Neeraj Agarwal with regards to the SPARTAN and TITAN studies looking at the depth of prostate specific antigen in those trials. The main result, or the conclusion, is that deep and rapid PSA response with apalutamide was really associated with a prolonged time to deterioration of health-related quality of life. Those questionnaires like FACT-P, physical wellbeing, BPI-sf, worse pain intensity and so on, fatigue. So this treatment intensification in M0 CRPC and also in mHSPC when you see, and this is good for us as treating urologists, the PSA is deep, this is a good sign with regards to health-related quality of life and also, as we know, with regards to prognosis. So we are satisfied as treating physicians and the patient always likes a very, very, deep, understandable, very deep PSA.

EE: Thank you for that and in those trials the patients did not know the PSAs. I don’t know if they snuck out and went and had them. But on top of those results think about knowing your PSA is undetectable. Great. Thank you all for joining and thank you for watching us and we’ll soon come back with more controversy and data.